The Female Happiness Collapse

An Essay on Declining Female Happiness and the Chemical Variable No One Controls For

In the 1970s, women in the United States reported higher happiness than men. By the mid-2000s, that gap had reversed. By 2024, across every available metric — happiness, life satisfaction, anxiety, depression, sadness, loneliness, anger, restless sleep — women scored worse than men in country after country, dataset after dataset, decade after decade.¹

During those same fifty years, women’s educational attainment surpassed men’s. The wage gap narrowed. Female workforce participation hit record levels. Legal rights expanded. Reproductive control became widely accessible. By every indicator that policymakers use to certify progress, women’s lives improved enormously.²

Three peer-reviewed studies, published between 2009 and 2024, track what happened to the experience of those lives while the indicators kept improving.

The Paradox Identified

Betsey Stevenson and Justin Wolfers, both economists at the Wharton School, assembled data from the General Social Survey (1972–2006), the Eurobarometer, the World Values Survey, the Monitoring the Future survey of American high schoolers, and the Virginia Slims Survey of American Women. Across every source, the pattern was the same: women’s subjective wellbeing had declined, both in absolute terms and relative to men.²

The decline was not confined to one demographic group, one survey instrument, or one country. It appeared across age brackets, education levels, racial categories, and industrialised nations. It was broad-based, persistent, and running in the opposite direction from every objective measure of women’s advancement.²

Stevenson and Wolfers laid out the standard explanations. The “second shift” — women entering the workforce without shedding household duties, producing a net increase in total burden. Arlie Hochschild described this in 1989. But time-use surveys complicated the story: total work hours had declined for both sexes since 1965, with women’s increase in paid work offset by reductions in unpaid domestic labour.²

Alan Krueger’s 2007 analysis of daily activities cut closer to the problem. He measured the pleasantness of how people actually spend their time. Women’s new activity mix — more market work, less home production — left them hedonically flat. Men’s mix shifted toward more enjoyable activities. Between 1966 and 2005, relative to men, women became hedonically worse off.² Not because their hours got longer, but because their days got worse.

Stevenson and Wolfers also raised a possibility that cuts deeper than workload. If women in the 1970s compared themselves to other women, the advances of the era felt like collective progress. As workplace integration increased, the comparison group shifted. Women began measuring their lives against men’s — and discovered the remaining gap. Greater equality made the remaining inequality more visible.² The women’s movement may have raised expectations faster than any society was prepared to meet them.

The paper left the paradox unresolved. But the data was unambiguous. Over thirty-five years, by every objective measure, women’s lives improved. When asked how they felt, women said: worse.

Support This Work

This work remains free because paid subscribers make it possible. If you find value here, consider joining them.

Paid subscribers get access to all books — including The DMSO Book, The Kitchen Remedies Guide, Chlorine Dioxide, The PSA Trap, Breast Cancer, and more — with 1-2 new books added each month. Plus the Deep Dive Audio Library: 180+ in-depth audio book summaries and discussions.

Pricing Update: The annual subscription moves from $50 AUD to $50 USD on May 1 — the first change in five years. Current paid subscribers keep their existing rate. Free subscribers can lock in the current price by upgrading before May 1.

The Paradox Complicated

Twelve years later, Leonardo Becchetti and Gianluigi Conzo at the University of Rome Tor Vergata pulled data from seven waves of the European Social Survey — 31 countries, nearly 200,000 observations — and added a dimension the earlier work had missed.³

After controlling for age, education, employment, marital status, income, health, and country, being female was associated with both higher life satisfaction and higher rates of depression. Women were more likely to rate their lives favourably. They were also more likely to report depressive episodes. Both findings were statistically significant. Both appeared in the same population at the same time.³

The finding was not fragile. The paradox held when the authors split the data by age group, education level, health status, geographic region, and survey wave. It held with and without Eastern European countries in the sample. It was structural.³

Becchetti and Conzo proposed a mechanism. They tested the “affect intensity” hypothesis — that women experience stronger emotional responses to both positive and negative events.³ The data confirmed it. Women’s satisfaction reacted more powerfully to good events — and more powerfully to bad ones. The emotional bandwidth was wider in both directions.

They also found that women are slower to recover from negative shocks — less likely to return to baseline happiness after an adverse event.³ Not because women are fragile. Because a life structured around chronic, low-level stressors — financial precarity, caregiving logistics, workplace friction, health management, the invisible coordination that holds households together — never stops generating shocks. Recovery requires a pause in the assault. The assault does not pause.

The profile of women who fell on the depressed side rather than the satisfied side was specific. They were younger, had less education, reported poorer health, and struggled to cope on their household income.³ These variables describe women squeezed hardest between the promise of progress and its actual delivery — women for whom the gap between what they were told their lives should look like and what those lives actually contain is widest.

Becchetti and Conzo reframed the question. The problem was not that women were less happy. It was that the same woman could report high satisfaction with her life while also reporting depression — and that this was not a contradiction but a feature of how women process the conditions they live in. They feel more of everything. The conditions supply plenty to feel bad about.

The Paradox Resolved

In February 2024, David Blanchflower of Dartmouth College and Alex Bryson of University College London published what amounts to the final report.¹

They assembled the most comprehensive dataset in the literature: the General Social Survey (1972–2021), the Eurobarometer (1975–2021), the European Quality of Life Surveys (2003–2016), the Gallup World Poll (2005–2021), the European Social Surveys (2002–2018), the UK Annual Population Survey (2012–2021), the Health Survey for England (2010–2019), and four British birth cohort studies. They tested whether the female happiness paradox — women reporting higher happiness alongside worse mental health — still existed.

It did not. The life satisfaction advantage that had made the pattern a “paradox” was gone. Men now reported higher happiness and higher life satisfaction. Women had lower wellbeing on every dimension.¹

The European Social Survey data across 39 countries laid it out. For the period 2002–2018, men were more satisfied with democracy, the economy, education, health services, and their standard of living.¹ On negative affect — depression, anxiety, sadness, loneliness, restless sleep — women scored worse on every measure, every one statistically significant.¹

The European Quality of Life Surveys from 36 countries replicated the finding with different questions. Men scored higher on cheerfulness, calm, vigour, restfulness, and daily engagement. Women scored higher on loneliness, downheartedness, tension, and feeling left out of society.¹

The old objection — that women simply report emotions more readily while men suppress — does not survive scrutiny. Oksuzyan and colleagues tested this directly in 2019 using the Survey of Health, Ageing, and Retirement in Europe. They found no gender-specific patterns in reporting behaviour.¹ The gap in the data tracks a gap in experience, not a gap in willingness to talk about it.

The events of 2020–2021 worsened the picture but did not create it. Women bore disproportionate burdens during the lockdowns and restrictions — school closures fell on them, they were overrepresented among furloughed and frontline workers — and their wellbeing declined faster.¹ But the pre-restriction data already showed the trends. The disruption accelerated a deterioration that was well underway before 2020.

The UK analysis went deepest. Using the Annual Population Survey, the Health Survey for England, and the birth cohort studies, Blanchflower and Bryson tested happiness, life satisfaction, anxiety, sadness, loneliness, enjoyment, helplessness, fearfulness, tension, and stress. Women scored worse on all of them.¹ The effect sizes were not marginal.

Fifteen years after Stevenson and Wolfers identified the paradox, it was gone. Not because women caught up. Because the last metric where they still held an advantage — life satisfaction — finally collapsed.

Where the Light Wasn’t

Three research teams. Fifteen years. Dozens of countries. Hundreds of thousands of observations. The General Social Survey, the Eurobarometer, the European Social Survey, the Gallup World Poll, the European Quality of Life Surveys, UK population surveys, birth cohort studies. They measured happiness, life satisfaction, anxiety, depression, sadness, loneliness, anger, sleep quality, cheerfulness, calm, vigour, tension, helplessness, fearfulness, stress, and feeling left out of society.

Not one controlled for hormonal contraception.

Approximately 82% of American women have used the oral contraceptive pill at some point in their lives.⁴ Over 60% of women of reproductive age currently use some form of hormonal contraception.⁴ The Pill arrived in 1960 and became the most commonly used contraceptive in America within a decade — the exact period when Stevenson and Wolfers’ data begins its downward turn.

The Pill’s psychological effects are documented in peer-reviewed research, though that research sits in pharmacology and neuroendocrinology journals that economists do not read. Sarah Hill, an evolutionary psychologist at Texas Christian University, describes a colleague excluding Pill users from his stress study because, although they reported feeling stressed, they showed no cortisol response — no physiological stress reaction at all.⁵ Multiple studies have confirmed this: women on the Pill show blunted or absent cortisol responses to standardised stressors.⁶ ⁷ ⁸ They show flattened daily cortisol rhythms.⁹ Research by Hertel and colleagues found Pill users exhibited biological markers typically associated with chronic stress or trauma — patterns suggesting the HPA axis, the body’s central stress-regulation system, is attempting to shut itself down under sustained artificial activation.¹⁰

Cortisol governs stress response, emotional regulation, learning, and memory encoding. The hippocampus, critical for memory, is vulnerable to chronic cortisol disruption; reduced hippocampal volume has been documented in Pill users.¹⁰ Hill describes her own experience of stopping the Pill as going from “grayscale to full color.”⁵

Becchetti and Conzo’s central finding — that women experience more intense emotional responses and are slower to recover from negative shocks — looks different against this evidence.³ They proposed “affect intensity” as an innate female trait. They never tested whether the affect patterns they measured — in a sample saturated with current and former Pill users — might be partly pharmaceutical in origin. The blunted stress response, the flattened cortisol rhythm, the chronic-stress biomarkers documented in Pill users describe a system under sustained chemical load. A system that has lost the capacity for normal recovery. A system that, when the chemical suppression lifts or fails, floods with unprocessed distress.

Blanchflower and Bryson’s data reads differently against this background. Women score worse on anxiety, depression, loneliness, restless sleep, tension, helplessness, and feeling left out of society.¹ These are precisely the domains governed by the HPA axis and cortisol regulation — the systems the Pill disrupts.

The Pill also alters mate selection. Research by Little and colleagues demonstrated that women who started hormonal contraception shifted their preferences toward less masculine male faces — a shift that did not occur in controls.¹¹ Roberts and colleagues found that women who chose partners while on the Pill reported lower sexual satisfaction but higher satisfaction with non-sexual partnership qualities — they chose reliable providers over sexually attractive partners.¹² When these women stopped the Pill, the preferences that guided their partner choice no longer matched their unmedicated desires. Stevenson and Wolfers noted that women have become less happy with their marriages over time, and that women initiate approximately 70% of divorces.² They attributed this to raised expectations. They did not consider that millions of women chose partners under chemical conditions that no longer applied.

This research is published and peer-reviewed. None of it appears in any of the three papers that mapped declining female happiness. The economists looked where economists look — at income, education, workforce participation, marital status. The pharmacological variable that altered the psychology of 82% of their study population was not in their models because it was not in their field of vision.¹⁵

Chemical Feminism

There is a second absence, harder to source but no less visible in the data.

The feminist life script that the Pill enabled — delay marriage, invest in education, build a career, have children later if at all — restructured the sequence of women’s lives. Sociologist Mark Regnerus has documented what followed.¹³ The Pill severed the link between sex and commitment. Men no longer needed to offer marriage to obtain sex. Women no longer needed marriage as insurance against pregnancy. The mating market split. Age at first marriage rose from the early twenties in 1960 to the late twenties. Cohabitation replaced marriage as the default. Women seeking commitment found themselves competing against women willing to accept less.¹³

The women most affected were those who most thoroughly followed the script. By their mid-thirties — educated, accomplished, career-established — the mating market had shifted against them. The delay that was supposed to optimise women’s lives optimised them for everything except the formation of stable families.

Stevenson and Wolfers found the steepest happiness declines among women with “some college” — women who partially followed the script.² Blanchflower and Bryson noted that women were less satisfied with their lives across every domain, including domains with no obvious economic explanation — loneliness, feeling left out of society, restless sleep.¹ These are not the signatures of insufficient career advancement. They are the signatures of disrupted connection — the downstream consequences of a mating market restructured by a pharmaceutical product, operating on a population whose mate preferences were chemically altered by that same product.

At the end of this cascade sits a question no study has examined: what is the psychological cost of delaying or forgoing motherhood? Not as a lifestyle choice assessed against career outcomes. As a biological drive, suppressed or deferred, in a population whose hormonal environment was pharmaceutically managed for decades. The Pill suppresses ovulation by mimicking pregnancy. The feminist life script delays actual pregnancy by design. The studies that would measure what this costs — longitudinal tracking of women’s wellbeing as a function of motherhood timing, controlling for desire, not just outcome — do not exist. The data that would answer the question was never collected.

And the women who do become mothers enter a medical system that pathologises the experience from the first prenatal appointment through permanent postpartum surveillance — 123 documented interventions that systematically replace a woman’s confidence in her own body with dependence on professional oversight.¹⁴

Three teams of researchers, working independently across fifteen years, exposed a pattern of declining female wellbeing so robust it appears in every dataset, every country, every metric. They searched thoroughly and competently in the areas where their training and funding directed them. They produced an important finding. They did not produce an explanation — because the variables that would generate one lay outside the illuminated zone.

This is the streetlight effect.¹⁶ Three teams searched where the light was. The keys were somewhere else.

The Streetlight Effect

Krueger’s hedonic analysis captured something the employment statistics missed entirely: between 1966 and 2005, the pleasantness of women’s daily experience declined relative to men’s.² Not their hours. Not their wages. The felt quality of being alive on a Tuesday afternoon. No government tracks this. No international ranking captures it. No politician campaigns on it.

The Gallup data covers 166 countries. The European surveys span 39 nations. The General Social Survey runs half a century. The sample sizes reach hundreds of thousands. The finding replicates across instruments, cultures, decades, and continents. Women’s lives got measurably better by every metric that institutions know how to track — and measurably worse by the metrics that measure what it is actually like to live them.

The studies that would connect these findings to the chemical intervention that 82% of women underwent — the longitudinal research tracking personality, stress response, mate selection, relationship satisfaction, and life outcomes before and after initiating hormonal contraception — have not been funded. The pharmaceutical industry has no incentive to discover that its most widely used product alters female psychology. Medical institutions prefer not to complicate contraceptive counselling. Advocacy organisations resist research that might be used to restrict access. The result is a gap where knowledge should be. Not an accidental gap. A structurally maintained one.

We have conducted a sixty-year experiment in population-scale hormonal intervention without tracking its psychological consequences, while simultaneously measuring the declining happiness of the population we intervened on — and never connecting the two. The economists who documented the decline didn’t think to look at pharmacology. The pharmacologists who documented the Pill’s effects didn’t think to look at happiness data. The gap between these two bodies of evidence is where the answer likely sits. No one is funded to stand in that gap.

If the most “progressed” population in the industrialised world is also the most anxious, most depressed, most lonely, and most sleepless — the problem is not with the population.

The problem is with what we decided to call progress. And with what we decided not to study about how that progress was chemically achieved.

---

References

¹ Blanchflower, D.G. & Bryson, A. (2024). The female happiness paradox. Journal of Population Economics, 37:16. https://doi.org/10.1007/s00148-024-00981-5

² Stevenson, B. & Wolfers, J. (2009). The paradox of declining female happiness. NBER Working Paper No. 14969. National Bureau of Economic Research.

³ Becchetti, L. & Conzo, G. (2022). The gender life satisfaction/depression paradox. Social Indicators Research, 160:35–113. https://doi.org/10.1007/s11205-021-02740-5

⁴ Daniels, K. & Abdo, M. (2022). Current contraceptive status among women aged 15–49: United States, 2017–2019. NCHS Data Brief No. 388.

⁵ Hill, S.E. (2019). This Is Your Brain on Birth Control. Avery.

⁶ Kirschbaum, C. et al. (1999). Impact of gender, menstrual cycle phase, and oral contraceptives on the activity of the hypothalamus-pituitary-adrenal axis. Psychosomatic Medicine, 61(2), 154–162.

⁷ Roche, D.J. et al. (2013). Hormonal contraceptive use diminishes salivary cortisol response to psychosocial stress and naltrexone in healthy women. Pharmacology Biochemistry and Behavior, 109, 84–90.

⁸ Bouma, E.M. et al. (2009). Adolescents’ cortisol responses to awakening and social stress. Psychoneuroendocrinology, 34(6), 884–893.

⁹ Meulenberg, P. & Hofman, J. (1990). The effect of oral contraceptive use and pregnancy on the daily rhythm of cortisol and cortisone. Clinica Chimica Acta, 190(3), 211–221.

¹⁰ Hertel, J. et al. (2017). Evidence for stress-like alterations in the HPA-axis in women taking oral contraceptives. Scientific Reports, 7(1), 14111.

¹¹ Little, A.C. et al. (2013). Oral contraceptive use in women changes preferences for male facial masculinity. Psychoneuroendocrinology, 38(9), 1777–1785.

¹² Roberts, S.C. et al. (2012). Relationship satisfaction and outcome in women who meet their partner while using oral contraception. Proceedings of the Royal Society B, 279(1732), 1430–1436.

¹³ Regnerus, M. (2017). Cheap Sex: The Transformation of Men, Marriage, and Monogamy. Oxford University Press.

¹⁴ Unbekoming. (2025). The Mother Who Remains: How Medicine Captures Women After Birth (Part 8). Lies are Unbekoming. Part of the eight-part Medicalized Motherhood series documenting 123 interventions across six phases of reproduction.

¹⁵ Unbekoming. (2026). Chemical Feminism. Lies are Unbekoming. Full treatment of the Pill’s documented effects on stress response, mate selection, and the feedback loops between hormonal contraception and feminist life patterns.

¹⁶ Unbekoming. (2026). The Streetlight Effect. Lies are Unbekoming. On the institutional architecture that ensures certain questions are never asked and certain studies never funded.